Content Tree Word Embedding for document representation

作者：

Highlights：

• A new approach is proposed in order to improve two main Deep Learning word embedding models.

• Previous document representations methods are reviewed.

• For the first time, a Content Tree-based approach is used in word embedding.

• Two classification tasks are performed using the presented approach.

摘要

•A new approach is proposed in order to improve two main Deep Learning word embedding models.•Previous document representations methods are reviewed.•For the first time, a Content Tree-based approach is used in word embedding.•Two classification tasks are performed using the presented approach.

论文关键词：Word embedding,Content tree,Word2Vec,GloVe,Sentiment analysis,Deep learning

论文评审过程：Received 27 September 2016, Revised 10 August 2017, Accepted 11 August 2017, Available online 16 August 2017, Version of Record 23 August 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.08.021